2857 (18) M. Blumenstein and M. A. Raftery, Biochemistry, 11, 1643 (1972). (19) B. Blrdsall, N. J. M. Birdsall. and J. Feeney, J. Chem. Soc., Chem. Common., 1972, 316 (1972). (20) W. Egan, S. Forsen, and J. Jacobus, J. Chem. Soc., Chem. Commun., 1972, 42 (1972). (21) R. H. Sarma and R . J. Mynott, J. Chem. Soc., Chem. Commun., 1972, 975 (1972). (22) R. H. Sarm and R. J. Mynon, J. Am. Chem. Sac., 95, 1641 (1973). (23) . , R. H. Sarma. R. J. Mvnott, F. E. Hruska. and D. J. Wood, Can. J. Chem., 51, 1843 (1973). (24) R. H. Sarma and R. J. Mynott, J. Am. Chem. Soc., 95, 7470 (1973). (25) M. Blumenstein and M. A. Raftery, Biochemistry, 12, 3585 (1973). (26) An equivalent, but less covenient, definition of the syn and anti forms has been put forth by M. Sundaraiingam, Biopolymers, 7,821 (1969). (27) J. H. Noggle and R. E. Schirmer, "The Nuclear Overhauser Effect Chemical Applications". Academic Press, New York. N.Y., 1971. (28) R. Freeman and H. D. W. Hill, J. Chem. Phys., 53, 5103 (1970). (29) T. A. Bryson, J. C. Wisowaty, R. B. Dunlap, R. R. Fisher, and P. D. Ellis, J. Org. Chem., 39,3436 (1974). (30) K. Isagawa, M. Kawai, and Y. Fushizaki, Nippon Kagaku Zasshi, 88, 553 (1967). (31) N. 0. Kaplan and M. M. Ciotti, J. Bioi. Chem., 221, 823 (1956). (32) A. M. Stein, J. K. Lee, C. D. Anderson, and B. M. Anderson, BiOCh8miSfry, 2, 1015 (1963). (33) The chemical shift anisotropy mechanism, (TicsA)-'. is usually zero since the magnitude of the chemical shift range for 'H's is quite small. The scalar coupling mechanism, (TiSC)-'. can be discounted because compared to YH. For further details see of the smaii value of y N for ref 27, p 21. (34) T. C. Farrar and E. D. Becker, "Pulse and Fourier Transform NMR lntro-
duction to Theory and Applications", Academic Press, New York, N.Y., 1971, p 46. (35) P. L. Johnson, C. L. Maier, and I. C. Paul, J. Am. Chem. Soc., 95, 5370 (1973). (36) P. L. Johnson, J. K. Frank, and I. C. Paul, J. Am. Chem. SOC.,95, 5377 (1973). (37) J. K. Lrank, N. N. Thayer, and I. C. Paul, J. Am. Chem. Soc., 95, 5386 (1973). (38) Reference 34, p 57. (39) G. A. Jeffrey, Acc. Chem. Res., 2, 344 (1969). (40) R. E. Schirmer. J. P. Davis, and J. H. Noggle, J. Am. Chem. Soc., 94, 2561 (1972). (41) A. P. Zens, P. J. Fogle, R. B. Dunlap, T. A. Bryson, P. D. Ellis, and R . R. Fisher, unpublished results. (42) Similar trends have been independently observed by Kaplan and coworkers, personal communication. (43) W. D. Hamili, Jr., R. J. Pugmire, and D. M. Grant, J. Am. Chem. Soc., 98, 2885 (1974). (44) T. J. Williams, R. M. Riddle, P. D. Ellis, R. R. Fisher, T. A. Bryson, and R . B. Dunlap, unpublished results. (45) T. J. Williams, A. P. Zens, P. D. Ellis, R. R. Fisher, T. A. Bryson. and R. B. Dunlap, unpublished results. (46) This argument is just a generalization of that put forth in ref 4. (47) D. G. Cross and H. F. Fisher, Biochemistry, 8, 1147 (1969). (48) G. Weber, Nature (London),180, 1409 (1957). (49) S.F. Velick in "Light and Life", W. D. McElroy and B. Glass, Ed., Johns Hopkins Press, Baltimore, Md.. 1961. (50) P. Walter and N. 0. Kapian. J. Biol. Chem., 238, 2323 (1963). (51) C. Sandorfy, "Electronic Spectra and Quantum Chemistry", PrenticeHall, Englewood Cliffs, N.J., 1964, p 4.
Synthesis and Biological Activity of enantio-[5-Valine]malformin, a Palindrome Peptide Miklos Bodanszky," Glenn L. Stahl, and Roy W. Curtis Contributionfrom the Department of Chemistry, Case Western Reserve University, Cleveland, Ohio 44106, and Department of Botany and Plant Pathology, Purdue University, West Lafayette, Indiana 47907. Received December 19, 1974
Abstract: The cyclic pentapeptide disulfide enantio-[5-~aline]malformin (11) was prepared by stepwise synthesis in solution. The ring was closed by the azide method and the dithiol was oxidized to the disulfide. The product is the enantiomer of a palindrome peptide, [5-valine]malformin; it has one-tenth of the biological activity of natural malformin in causing curvatures on corn roots, inhibiting adventitious root formation, and stimulating the growth of etiolated bean cuttings. It induces malformations on bean seedlings which are identical with those induced by the parent compound.
Malformins, a family of metabolic products of Aspergillus niger, induce malformations on higher plants,' inhibit adventitious root formation,2 have antibacterial3 and cytotoxic properties, and under some conditions stimulate plant g r ~ w t h .The ~ initially proposed5 structure for malformin A I , a principal member of the family hereafter referred to as malformin, was recently The revised sequence, I, was confirmed by synthesis.' D-Leu /
Ilej
4
The decision to synthesize enantio- [ 5-valine] malforming was based on the following considerations. In some biologically active members of the malformin family, isoleucine is replaced by valine.I0 Although the sequence of these constituents has not been determined, it seemed to be likely that there are no additional structural differences between malformin and other members of the family, or if there are differences, they do not affect the biological activity in any major way. Therefore, we expected that the enantio analog of [5-valine]malformin (11) would be a suitable model for a
\
3Val
/Leu\ D-Val D-Vd \
I The examination of structure-activity relationships in malformins, including a study of their different stable conformations,* necessitated additional efforts toward the synthesis of malformin analogs. This work, however, was hampered by the limited availability of one of the constituent amino acids, D-cystine. Therefore, we decided to prepare and study the enantio form of a malformin.
* Address correspondence to this author at Case Western Reserve University.
/
cys-cys U
I1
study of the conformation of malformins and also hoped that the synthetic material would be biologically active. This hope was supported by the antibacterial properties of retro-enantio analogs of gramicidin SI1 and of enniatins.l* T h e t o p ~ c h e r n i c a similarity l~~ between a microbial peptide and its enantiomer is further enhanced if the direction of the chain is reversed.14 Such retro enantiomers are closely Bodanszky, Stahl, Curtis
/
enantio-[5- Valinelmalformin
2858
-
I
I
I
I
16
I
I
-
I I-\ \
8-
I
4-
I I
-4-
-12
-
I
-16
-
I I
I
I
I
I
I
*/--
I
I
I
240
d
\
\
-
'--e
I
I\ / I "
I
-\
$ 1
200
I
,\:
I \ I
I I
I
-
\;
\
1
! l
-20-
I
CD
-
\
L
x
-8
\
I
b o E
-
ORD
N
U
I
-
20
12
I
I
I
I
280
I
I
320
I
I
similar to their parent molecules and should behave similarly in many systems, e.g., in forming complexes with metal ions. The antibacterial potency of enantio and retro-enantio analogs of peptide antibiotics is probably related to their similar i ~ n o p h o r e properties. '~ On the other hand, a cyclic peptide and its retro enantiomer are certainly not identical and they might be distinguished by some biological receptors. Thus, the expectation that on different plants enantio[ 5-valine]malformin should exert the various effects of natural malformin had to be proved or disproved by experiments. It was our intention to maintain topochemical similarity between the synthetic analog and malformin. In this respect it was not obvious whether L-isoleucine should be replaced by D-isoleucine or D-alloisoleucine in the enantiomer. Replacement of isoleucine by valine eliminated this problem and a t the same time resulted in a palindrome peptide, a sequence which is identical with its retro sequence. It was not unreasonable to assume close topochemical similarity between malformin (I) and the planned symmetrical analog 11. The synthesis of I1 was carried out similarly to that of by stepwise chain lengthening from the C-terminal residue.16 The protected linear intermediate, benzyloxycarbonyl-D-valyl-S-benzyl-L-cysteinyl-Sbenzyl-L-cysteinylD-valyl-L-leucine methyl ester, was converted to the hydrazide, partially deprotected with HBr in acetic acid, and cyclized via the azide17 to the S,S'-dibenzyl derivative of 11. After removal of the S-benzyl protecting groups by reduction with sodium in liquid ammonia,Is the dithiol was heated in dimethyl sulfoxide to convert it to the more active conformation8 and then oxidized to the disulfide with iodineIg or by diiodoethane.20After purification by chromatography on silica gel,Io 11 was secured in crystalline and analytically pure form. It travelled as a single spot on thin-layer chromatograms and exhibited ORD-CD spectra (Figure 1) which were closely similar to the mirror image of the spectra of I (cf. ref 7 ) . The biological activity of malformin (I) and enantio-[5valine]malformin (11) was compared in a series of tests and confirmed our expectations. In the corn root curvature assay,' the optimum concentration of malformin (I) and en1 6 3 7
Journal of the American Chemical Society
I
3 6 0 200
I
I
240
I
I
280
I
I
320
I
I
36C
antio-[5-valine]malformin (11) was 2 X lo-' and 2 X M , respectively (Table I). The concentration M ) of synthetic I1 required to stimulate the growth of etiolated cuttings of Phaseolus aureus4 was tenfold greater than that required for similar responses to malformin (Table I). In addition, the concentration M ) of synthetic I1 required to inhibit adventitious root formation2 of similar cuttings of P.aureus was tenfold greater than that required for similar responses to malformin. W e concluded that synthetic I1 is about one-tenth as active as natural malformin. In a qualitative test using seedlings of P. vulgaris, the synthetic product induced malformations which could not be distinguished from those induced by malformin itself.'
Experimental Section Capillary melting points are reported uncorrected. For thinlayer chromatography, the following systems were used: A, CHCI3-CH30H (l9:l); B, CHC13-CH30H (9:l); C, CHC13T F E (85:15); D, water-saturated ethyl acetate. Spots on T L C were revealed with tert-butyl hypochlorite-KI-starch reagents.2' For amino acid analysis, samples were hydrolyzed in a mixture of concentrated HCI and HOAc (1:l) at 110' for 16 hr and analyzed by the Spackman-Stein-Moore method22 on a Beckman Spinco 120C amino acid analyzer. The following abbreviations are used: DMF, dimethylformamide; DMSO, dimethyl sulfoxide; TFA, trifluoroacetic acid; DIPEA, diisopropylethylamine; T H F , tetrahydrofuran; TFE, 2,2,2-trifluoroethanoI. Benzyloxycarbonyl-D-valyl-L-leucinemethyl ester (111) was prepared similarly to its e n a n t i ~ m e r :yield, ~ 98%; mp 134-13s'; (c 2, DMF); T L C R/ 0.63 [lit.7 mp 135-136'. [ c ? ] ~ ~-25.5' D [c?]~~ +25.4' D (c 2, DMF)]. N-Benzyloxycarbonyl-S-benzyl-L-cysteinyl-D-valyl-L-leucine methyl ester (IV) was prepared analogously to its enantiomer:' (c 2, DMF); T L C Rf yield, 73%; mp 160-161'; [ c ? ] ~ -42.1' ~D 0.68 [lit.7 mp 160-161': [aI2'D +40.0' (c 2, DMF)]. N-BenzyloxycarbonyI-S-benzyl-L-cysteinyl-S-benzyl-L-cysteinyl-D-valyl-L-leucinemethyl ester (V) was prepared by the method used for its e n a n t i ~ m e r :yield, ~ 73%; mp 190-191'; [ c Y ] ~ ~ D -44.4' (c 2, DMF); TLC Rf 0.68 [lit7 mp 193-194'; [ C Y ] ~ ' D +44.9' (c 2, DMF)]; amino acid analysis, Val 1.00; Leu 1.00, Cys(Bz1) 2.00. Anal. Calcd for C ~ O H ~ ~ O(765.0): ~ N ~ C, S Z62.8; H , 6.8; N , 7.3. Found: C , 63.0; H, 6.9; N , 7.1. Benzyloxycarbonyl-D-valyI-S-benzyl-Lcysteinyl-S-benzyI-Lcysteinyl-D-valyl-L-leucineMethyl Ester (VI). A sample of V (1 3.2
/ 97:lO / M a y 14, 1975
2859 Table I. Biological Activity of enantio- [ 5-Valine] malformin and Malformin ~
~~
~
Concn, M
enanrio[ 5-Valine] malformin
Corn root curvature assay' 2x 2x 2 x lo-' 2 x lo-'
H,O (control) Phaseolus aureus growth assay4 1o - ~
lo-'
1o-8
H,O (control)
-
Malformin
% roots with >90" curvature 18.2 49.0
33.7 77.9 25.0
15.2 9.2
Growth increments, cm 2.84 3.42a 2.72
3.02 3.360 2.60 3.04
asignificantly different from controls at 5% level.
g, 17 mmol) was suspended in 3 M HBr in HOAc (32 ml). After 2 hr, some material remained undissolved and therefore more HBr in acetic acid (8 ml) was added. After 1 hr, the mixture was diluted with ether (450 ml); the precipitated hydrobromide was collected, washed with ether (100 ml), and dried in vacuo over NaOH. The partially deprotected peptide derivative was dissolved in D M F (50 ml), and benzyloxycarbonyl-D-valine p-nitrophenyl ester23 was added, followed by DIPEA (3.1 ml). After 24 hr, the product began to crystallize. At the end of 48 hr, the reaction was complete. The crystals were filtered and washed with D M F (20 ml) and with methanol: yield, 10.2 g. The washings were pooled with the filtrate, and the solvent was removed under reduced pressure. The residue was dissolved in CHCI3 (100 ml) and washed with water (50 ml) and 0.1 M HCI (10 ml). The solvent was removed in vacuo and the solid material triturated and washed with ether (100 ml): yield, 1.5 g. The two crops were pooled and washed with EtOAc: yield, 11.5 g (78%); mp 222-224'; [a]25D -41.1' (c 2, DMF); TLC, a single spot at R p 0.57; amino acid analysis, Val 2.05; Leu 1.OO,Cys(Bz1) 1.95. Anal. Calcd for C&6108N& (864.1): C, 62.5; H, 7.1; N , 8.1. Found: C, 62.3; H, 7.1; N , 7.9. cycfo-(S-Benzyl-L-cysteinyl-Sbenzyl-L-cysteinyl-D-valyl-L-leucyl-Svalyl) (VII). A sample of VI (1 .O g) was suspended in a mixture of peroxide-free T H F (50 ml), hydrazine (5 ml), and methanol (6 ml). Dissolution occurred within 5 min; after 15 min, the product began to precipitate. After standing overnight at room temperature, the product was collected by filtration and washed with T H F and methanol: yield, 0.84 g (84%); mp 251-252'; [a]25D -16.2O (c 2, TFA); TLC RB 0.54. A sample of the hydrazide (1.82 g, 2.1 mmol) was partially deprotected by treatment with 3 M HBr in acetic acid (20 mi) for 90 min. The hydrobromide was precipitated with ether (400 ml), washed with ether, and dried in vacuo over N a O H and P20s: yield, 1.76 g. It was dissolved in D M F (20 ml) and cooled to -20'. Sodium nitrate ( 1 M solution, 2.1 m]) was added to the copled mixture, which was stirred for 20 min, and then cold D M F (250 ml) was added. The solution was made alkaline with DIPEA ( 1 . 1 ml), set aside at 4O for 48 hr, and kept a t room temperature for 24 hr. Most of the solvent was removed under reduced pressure, and the cyclic material was precipitated with water (200 ml), collected on a filter, washed with warm ethanol, and dried: yield, 0.64 g (0.93 mmol, 43%); mp
>300', slight decomposition above 250'; [cu]"D -62.7' (c 2, TFA); T L C RF 0.51; amino acid analysis, Val 2.05, Leu 0.95, Cys(Bz1) 2.00. Anal. Calcd for C ~ ~ H S I O ~ (697.9): N S S ~ C, 61.9; H, 7.4; N, 10.0. Found: C, 61.7; H, 7.2; N, 10.3. enentio-[5-Vatine]malformin (11). A sample of VI1 (1 13 mg) was suspended in liquid NH3 (ca. 100 ml), and small pieces of sodiumI8 were added until the blue color persisted for 2 min. The color dispelled with "&I. The NH3 was allowed to evaporate under N2 and finally in vacuo. The residue was washed with 0.1% HOAc (2 ml) and dried: yield, 81.5 mg (98%). The dithiol was dissolved in DMSO (16 ml); the solution was warmed on a steam bath for 1 hr, cooled to room temperature, and titrated with an iodine solution (0.05 M in DMSO, 2.9 ml, calcd 3.1 ml). The small excess of iodine was reduced with Na2S203 (0.5 ml, 1% in H20). The mixture was diluted with ethyl acetate (150 ml) and washed with water (300 ml). The precipitate that formed8 was removed by centrifugation and the solvent by evaporation. The residue was triturated with ether. The product (33.5 mg, 41%) was purified by chromatography on silica gel.Io The purified material (90% recovery) gave a single spot on T L C (RP 0.44) and exhibited ORD-CD spectra that were quite similar to the mirror image of the spectra of malformin A (Figure I): amino acid analysis, Cys 1.95, Val 2.05, Leu 1.OO. Anal. Calcd for C22H3705N~Sl(515.7): C, 51.2; H, 7.2; N, 13.6. Found: 51.4; H, 7.2; N , 13.6.
Acknowledgments. Amino acid analyses were carried out by Mrs. Delores B. Gaut and elemental analyses by the Baron Consulting Co., Orange, Conn. This study was supported by grants from the U S . Public Health Service ( N I H AI-075 15) and the National Science Foundation (GB7 158). References and Notes (1)R. W. Curtis, Plant Physiol., 33, 17 (1958);Science, 128,661 (1958). (2)R. W. Curtis and G. Fellenberg, Plant CellPhysioL, 13,715 (1972). (3)S.Suda and R. W. Curtis, ADD/. Microbial., 14,475 (1966). (4)W. John and R. W. Curtis, Experientia. 30, 1392 (1974). (5) S.Marumo and R. W. Curtis, Phytochemistry, 1, 245 (1961). (6)M. Bodanszky and G. L. Stahl, Proc. Natl. Acad. Sci. U.S.A., 71, 2791 (1974). (7)M. Bodanszky and G. L. Stahl, Bioorg. Chem., in press. (8) K. Anzai and R. W. Curtis, Phytochemistry, 4, 713 (1965). (9)For the nomenclature of peptide analogs, see M. Bodanszky and V. du Vigneaud, J. Am. Chem. Soc.. 81, 1258 (1959), footnote 2. (IO) S.Takeuchi, M. Senn, R. W. Curtis, and F. W. McLafferty. Phytochemistry, 6, 287 (1967). (11)M. M. Shemyakin. Yu. A. Ovchinnikov. V. T. Ivanov, and I. D. Ryabova. Experientia, 23,326(1967). (12)M. M. Shemyakin, Yu. A. Ovchinnikov. and V. T. Ivanov, Angew. Chem., ht. Ed. Engl., 8,492 (1969). (13)Yu. A. Ovchinnikov, V. T. Ivanov, and M. M. Shemyakin, Pept., Proc. Eur. Pept. Symp., 8th, 1966. 173 (1967). (14)While this paper was under revision, we became aware of an additional example [T. Wleland, B. Penke, and C. Birr, Justus Liebigs Ann. Chem., 759,71 (1972)]illustrating this point. (15)B. C. Pressman, Antimicrob. Agents Chemother., 1969, 28 (1970). (16)M. Bodanszky, Ann. N.Y. Acad. Sci.. 88, 655 (1960). (17)T. Curtius, Ber., 35,3226 (1902). (18)V. du Vigneaud, L. F. Audrieth, and H. S.Loring, J. Am. Chem. SOC.,52,
4500 (1930). (19)K . Anzai and R. W. Curtis, Phytochemistry, 4, 263 (1965). (20) A. Schoberl, M. Rimplet, and E. Clauss, Chem. Ber., 103,3159(1970). (21)R. H. Mazur, B. W. Ellis, and P. Cammarata, J. Biol. Chem., 237, 1619 (1962). (22)0.H. Spackman, W. H. Stein, and S. Moore, Anal. Chem., 30, 1190 (1958). (23)Prepared according to Biochem. Prep., 9, 110 (1962).
Bodanszky. Stahl, Curtis
/
enantio-15- Valinelmalformin
